EP1960309B1 - Method for producing nanostructures on a substrate - Google Patents

Abstract

The method involves adding drops of solution for nanostructure formation in water at a temperature above a predetermined temperature, in which the drops of solution floats to a substrate on a steam pad. The steam pad heats the substrate to evaporate the added drops, leaving and forming nanostructures on the substrate. The temperature of the substrate is set above 200 degrees Celsius, and is tilted before or immediately after dropping the solution. The substrate is heated in several minutes or few hours for continuous growth of nanostructures.

Description

The invention relates to a method for producing nanostructures on a substrate. The invention relates in particular to the production of nanowires or linear arrangements of nanodots (clusters) and the production of carbon nanotubes directly on a largely arbitrary carrier. The invention also relates to the production of nanostructures on silicon wafers, in particular on MEMS or microchips.

Nanostructures such as nanowires and nanotubes are currently the focus of current research. They represent a class of materials that u. a. due to quantum effects have novel electrical, optical, magnetic and thermodynamic properties. In addition to the primarily academic questions, there is also the problem of reproducible mass production with the simplest possible means for the timely expansion of the field of commercial use of nanostructures.

Although simple methods for the production of nanoparticle aggregates are known (see, for example, US Pat Tsapis et al., "Onset of Buckling in Drying Droplets of Colloidal Suspension", Physical Review Letters, 94, 018302-1 - 018302-4, 2005 However, the generation of nanostructured nanostructures on the surface of a substrate is still a difficult process, usually associated with multiple steps and high costs. Typical processes for the preparation of such structures are Vapor Liquid Solid (VLS) or MOCV processes. While these methods are relatively universally applicable, both the control of the atmosphere (UHV) and the need for high temperatures (600-1000 ° C) require expensive equipment and make synthesis time consuming. Pre-structured substrates such as MEMS can not readily be exposed to such high temperatures.

To reduce the expense, the skilled worker also knows wet-chemical production process from aqueous solution, which lead to the desired results at temperatures below 100 C ° and at atmospheric pressure (eg see Law et al., "Nanowire dye-sensitized solar cells", nature materials, 4, 455-459, 2005 and Satoshi et al. "Growth conditions for wurzite zinc oxide film in aqueous solutions", J Mat. Chem, 12, 2002, pages 3773-3778 ). However, in addition to their very slow process (process times of several hours to days), wet-chemical methods have other disadvantages. For example, no epitaxial growth on silicon is possible (see J. Phys. Chem. B 2001, 105, 3350-3352 ). In addition, some solvents are also used, which bring a disposal problem with it.

For example, one is currently very interested in the production of zinc oxide (ZnO) nanostructures, such as nanorods and nanotubes. This is the case because ZnO, as a semiconductor, can form a large variety of nanostructures. Moreover, versatile applications as optoelectronic devices, lasers, field emission and gas sensor materials are considered (for the fabrication and application of nanotubes and nanorods, see also Advanced Materials 2005, 17, 2477 ). In order to epitaxially produce ZnO structures, either special substrates such as gallium nitride (GaN) are used or silicon substrates are coated with a so-called "seeding layer", which mostly consists of a ZnO thin film heated at 400 ° C. A direct non-epitaxial production of nanostructured nanostructured ZnO structures on substrates is not yet known.

Another example is carbon nanotubes (CNTs), which are called smart materials. These have their application in fuel cells, bio-gas sensors, field effect transistors. Again, the known manufacturing processes (arc, laser, CVD, PECVD) on a high technical effort (high temperature, vacuum, etc.). An alternative solvothermal synthesis process at low temperature (310 ° C) solves this problem ( Wang et al., Nanotechnology 16, 21-23, 2005 ), but the process still needs about 20-40 hours in low yield.

A third example is water-soluble nanostructures of inorganic materials such as CaC03, BaC03, which are recommended for new applications in biotechnology with unusual mechanical and optical properties. The controlled production of such materials is carried out by mixing salts with polymers (biopolymers, see Shu et al., Nature materials 4, 51, 2005 ). Wires of such material, which show a diameter of less than 100 nm, are not yet known.

An even more important example is the production of nanowires that consist of nanoclusters. It is already known that the arrangement of nanoclusters in 1D, 2D or 3D can lead to new properties that do not occur in disordered clusters and are based on the nearest interaction between the clusters, such as Magnitization Flip (application: data storage) and plasmonic conductivity (Application: optical fiber) (see eg as Sources: nature materials, 2, 229, 2003 or Eur. J. Inorg. Chem. 2455, 2001 ).

In contrast to the known and simple arrangement of nanoparticles in 2D and 3D, the 1D arrangement is a complicated process in which typically a template (eg mask, mold) must be used. This template limits the materials used and may result in disruption of the clusters Nanowire (for example, when removing the template) or its properties (eg, if a nanowire is used as a sensor, then not completely removed template material can reduce the sensitivity).

Thus, the various processes used to make nanostructures often have common drawbacks such as complexity, high cost, low speed.

It is therefore the object of the invention to specify a method for producing nanostructures on a largely arbitrary substrate, which leads to high yield and large-area coverage even after a very short process time without expensive equipment.

The object is achieved by a method having the features of the main claim. The subclaims indicate advantageous embodiments of the invention.

The method according to the invention differs from all methods of producing nanostructures known to the person skilled in the art in that it does not rely on a gradual, self-organized formation of these structures from their constituents, as is the case, for example, with epitaxial growth. Rather, a highly unstable state characterized by nonequilibrium reactions forms the starting point of nanowire growth in the process described herein. The invention makes use of a natural effect that is not fully understood until now. However, the method gives very good reproducible results and seems to work for a number of different materials, of which only a selection can of course be experimentally verified, but of course this is not intended to limit the invention.

1. 1. Einbringen eines eine Nanostruktur bildenden. Materiales in Wasser zur Erzeugung einer Lösung.
2. 2. Gegebenenfalls Zugabe von Katalysator-Partikeln in die Lösung.
3. 3. Aufheizen des Substrats auf Temperaturen oberhalb 200 °C.
4. 4. Zugabe einzelner Tropfen der Lösung auf das geheizte Substrat.
5. 5. Es erfolgt ein Verdampfen der Tropfen unter Nutzung des Leidenfrost-Effekts, bei dem sich Nanostrukturen bilden und auf dem Substrat abgeschieden werden.
6. 6. Gegebenenfalls wird das Substrat geneigt, so dass ein der Schwerkraft folgender, bewegter Tropfen Nanostrukturen entlang seines Weges deponiert.

The method according to the invention essentially comprises the steps:

1. 1. Introduction of a nanostructure forming. Material in water to produce a solution.
2. 2. Optionally, adding catalyst particles to the solution.
3. 3. heating the substrate to temperatures above 200 ° C.
4. 4. Add individual drops of the solution to the heated substrate.
5. 5. Evaporation of the drops takes place using the Leidenfrost effect, in which nanostructures form and are deposited on the substrate.
6. 6. Optionally, the substrate is tilted so that one of the gravitational forces of following, moving droplets deposit nanostructures along its path.

It seems necessary for the formation of the nanostructures that the solution evaporates locally explosively. In this case, the Leidenfrost effect seems to play an important role, according to which a drop of water falling on a hot plate floats on a pad of steam and slides, whereby its evaporation is delayed somewhat. The effect occurs when the hotplate has a temperature beyond the Leidenfrost temperature which is slightly above 200 ° C at normal pressure for water. Regardless of the question of the reaction kinetics, which is currently unclear, it is extremely advantageous, if not necessary, to exploit the Leidenfrost effect when dripping on the material solution. Because of the sliding of the droplet over the substrate, a uniform distribution of the deposited nanostructures is favored and possibly also covers a larger area.

The experimental results prove that the first nanostructure structures have formed over a large area - in the area of the deposited drop - within a few seconds. It is particularly noteworthy that the same process steps to produce very different structures that z. T. even at the same time, can serve. The various products of the method according to the invention will be illustrated below with reference to scanning electron micrographs and explained in more detail.

In Fig. 1 Zinc oxide nanorods can be seen, which are formed directly by the evaporation of drops of an aqueous 0.02 M zinc acetate solution. The heated silicon substrate does not require a "seeding layer" for this purpose and no catalysts are used. The structures shown are formed after 2-3 minutes of processing time and cover the substrate over a large area in the area previously wetted by the drop surface. ZnO is formed from the zinc ions of the solution and presumably the oxygen in the ambient air. Heating the substrate only at temperatures between 25 and 200 ° C, so no nanostructures, but only a thin film.

A preferred embodiment of the invention consists in the addition of catalyst particles, preferably of noble metal nanoparticles. More preferably, gold particles having a diameter of about 20 nm are added to the nanostructured aqueous solution described above.

The gold particles themselves are commercially available in solution and stabilized by organic and inorganic additives as a solution from Aldrich. The stabilization prevents the agglomeration of the particles and is indispensable. As it turns out, the carbonaceous stabilizers also tend more or less inevitably to the formation of nanostructures when using the method according to the invention.

Fig. 2 shows zinc oxide nanotubes that form after a few minutes due to the addition of catalyst particles (ZnO solution with Aldrich solution in the ratio 1: 3). Fig. 2 Below is a partial enlargement of the upper image to more clearly illustrate the tube structure. Again, the result shown can not be achieved if the substrate is heated below 200 ° C or heated only gradually.

An interesting side effect of the addition of the catalyst particles relates to the fate of the gold on the substrate. The gold can not escape from the hotplate when the water is drained, but it does not appear to be readily incorporated into the nanostructures mentioned so far. Rather, one finds that the gold, itself a fractal-seeming, but evenly distributed gold net forms on the substrate, which is still under other nanostructures (here: protruding wires, see below), such as Fig. 3 busy. Whether this network also plays a role in the formation of more complex nanostructures is still uncertain. The gold mesh can be produced, for example, with a solution of 0.03 M ZnO in 0.1 M NaOH while admixing the Aldrich solution in a volume ratio of 1: 6.

In Fig. 4 see transmission electron micrographs after using the same ZnO / NAOH / Aldrich solution in which a network of "multiwall carbon nanotubes" (MWCNT) can be seen (note the detail enlargements). Curiously, these MWCNT lie on the substrate right from the beginning and thus form an electrically conductive mesh. The carbon comes from the stabilizer of the Aldrich solution in the examples. However, the MWCNT can also be generated quite deliberately, for example when a pure carbonate solution drips onto the substrate.

If the substrate is allowed to rest under ambient air after being dripped with the previously mentioned solution, protruding wires (tufts) grow from the in Fig. 4 also shown crystals. In the course of an hour, the nanowires continue to grow, increasing the coverage density and decreasing the wire thickness. This seems to rejuvenate the existing wires, and new smaller diameter ones are formed. The wire diameters vary between initially 100 nm and later about 60 nm, while the length increases from about 3 μm after 15 minutes to 25 μm after one hour. Thereafter, a further growth of the wires is no longer discernible, although it does not have to come to a complete halt. Fig. 5 shows the nanowires growing away from the substrate at different stages of their formation. The times of the snapshots after applying the drops of solution are given next to the respective pictures. The vertical wires are inorganic and water soluble. They consist of carbon (from the stabilizer of the Aldrich solution), sodium (from the caustic soda to dissolve the ZnO) and oxygen (presumably from the ambient air).

It should be emphasized that in the examples given above, all nanostructures have been formed directly from ions in solution (except for the fractal gold mesh, which is formed from gold colloids). Thus, there is no need to pre-populate nanoparticles into the solution to deposit them on the substrate. This is a significant difference from conventional landfill e.g. of carbon nanotubes, which otherwise have to be in powder form before. Nevertheless, the process according to the invention can of course also be carried out with the addition of nanoparticles to the aqueous solution, which is then added dropwise.

A very advantageous embodiment of the method according to the invention consists in tilting the substrate against the horizontal before or immediately after the application of the solution drop.

By tilting the substrate, nanostructures can be deposited relatively selectively along a preferred direction (the slope gradient). The material to be formed or deposited is dissolved or suspended in water as before. As above, substrates are heated to temperatures above 200 ° C. Then droplets with diameters of about 1-2 mm are brought to the surface. In contrast to the previous one, the substrate is tilted, the tilt angle now determining the drop drainage speed. Due to the Leidenfrost effect, the droplet floats on a vapor cushion.

By delivering material from the drop, which takes place predominantly at the edge of the drop ("coffee stain effect"), nanodots (clusters) are deposited on the substrate along the direction of movement during the drop run. This happens at relatively regular intervals as a function of the running speed of the drop, that is to say from the set tilt angle. If the clusters are sufficiently dense, they will in effect form a nanowire, e.g. electrically conduct. Molecules can accumulate between the clusters, which could be useful in chemical sensor applications. However, there is also a sorting effect with regard to the cluster size: the largest or heaviest clusters are first deposited, the smaller ones later. Thus, along the direction of travel of the droplet, the cluster size distribution changes from large to small, exhibiting locally very small variation, i.e., small variations. Extensive sections of wire consist of roughly equal sized clusters.

The dumping landfill is in obvious contrast to the well-known lotus effect, in which weakly adhering material is collected by a passing water droplet and taken with it, which is why it could deserve the name "anti-lotus effect".

The temperature of the substrate above 200 ° C for enforcing the Leidenfrost effect also plays a major role here, since the sliding of the droplet on its own steam pad apparently has a favorable effect on the uniform distribution of the later detectable distribution of nanodots.

An example of the application of the anti-lotus effect, ie the combination of Leidenfrost effect, coffee-stain effect and tilting is in Fig. 6 to see. In this case, a drop of zinc acetate / water (just like ZnO nanorods) is used. The top picture shows parallel nanowires consisting of nanorods. The bottom shows a nanowire of nanodots separated from each other. In both cases, the nanostructures emerge within a few seconds after the ZnO solution drips.

Another example (not shown) relates to the preparation of silver structures. It is known that the thermal decomposition of AgNO ₃ (silver nitrate) to metallic silver takes place at 180 ° C, so that fabrication and arrangement of the nanoparticles in a 1D array can be realized in one step.

Apparently, many different materials can be produced easily and quickly in this way and without a template. Research to explore the potential of the invention is still in its infancy. However, it has been clear so far that steps 1, 3, 4 and 5 of the process described at the outset are the necessary prerequisite for a still largely unrecognized reaction kinetics in the formation of nanostructures on substrates, whereby the substrates themselves only have to be required to have temperatures withstand slightly above 200 ° C. The influence of the surface roughness of the substrate remains to be investigated. In the experiments presented here, monocrystalline silicon wafers were possibly used with an SiO ₂ covering layer.

Claims (6)

1. A method for producing nanostructures on a substrate,
   characterized in that
   a solution of nanostructure forming material in water is dripped onto said substrate at a temperature above the temperature at which a drop of the solution is initially suspended on a vapor cushion after being applied by dripping onto the substrate, with nanostructures being formed when the drops evaporate.
2. The method of claim 1, characterized in that the temperature of the substrate is adjusted to above 200 °C.
3. The method of one of the preceding claims, characterized in that the substrate is tilted against the horizontal plane before or immediately after having been dripped onto said substrate.
4. The method of one of the preceding claims, characterized in that the substrate is still heated between several minutes and a few hours after evaporation of the drop, so that distant nanowires continue growing.
5. The method of one of the preceding claims, characterized in that catalytic nanoparticles of noble metal are admixed to the solution.
6. The method of claim 5, characterized in that gold particles with a diameter of about 20 nm are admixed to the solution.

Images